Understanding User Adoption of Mobile Technology

2 downloads 0 Views 4MB Size Report
May 11, 2005 - from two telephone interviews in Finland', Int. J. Electronic Healthcare, ...... within the field (Checkland and Holwell, 1998). ...... 1996, Richard D.
Understanding User Adoption of Mobile Technology: Focusing on Physicians in Finland Second Edition

Shengnan Han

DOCTORAL DISSERTATION

To be presented with the permission of the Faculty of Economics and Social Sciences at Åbo Akademi University for public criticism in DataCity, entrance A, third floor, Auditorium 3102, on the 11th of May 2005, at 12 o’clock noon.

Åbo Akademi University Turku Centre for Computer Science/ Institute for Advanced Management Systems Research Lemminkäinengatan 14 B, Fin-20520 Åbo Finland

Tur ku Cent r e for Comput er Science Åbo, 2005

SUPERVISORS Prof. Dr. Pirkko Walden Department of Information Systems / Institute for Advanced Management Systems Research Åbo Akademi University Åbo, Finland Prof. Dr. Christer Carlsson Department of Information Systems / Institute for Advanced Management Systems Research Åbo Akademi University Åbo, Finland

REVIEWERS Prof. Dr. Per Egil Pedersen Department of Information and Communication Technology Agder University College Grimstad, Norway Prof. Dr. Virpi Kristiina Tuunainen Department of Business Technology, Information Systems Science Helsinki School of Economics and Business Administration Helsinki, Finland

OPPONENT Prof. Dr. Harry Bouwman Information and Communication Technology Faculty Technology, Policy and Management Delft University of Technology Delft, the Netherlands

ISBN 952-12-1523-2 ISSN 1239 -1883 Painosalama Oy, 2005 Turku, Finland

ABSTRACT Mobile technology is transforming our life. It improves as more people adopt it and gain experience of it. The primary objective of the dissertation is to contribute knowledge about physicians’ adoption of mobile technology and their intentions to use this technology in the health-care setting in Finland. With the aid of previous research into the adoption of information systems, mainly the technology acceptance model (TAM), the unified theory of acceptance and use of technology (UTAUT), personal innovativeness in the domain of IT (PIIT) and new concepts of individual differences in fragmentation of working time (FWT) and fragmentation of working space (FWS), two research models were proposed and empirically examined, using data collected in the Duodecim project, which involved 800 physicians who were working in the Finnish health-care sector in a pilot trial of a mobile medical information system from June of 2003 till the end of 2004. This system is operated with the Nokia Communicator 9210. The results suggested that the two proposed research models could adequately explain physicians’ intentions to use the mobile system. Perceived usefulness was the most strong and dominant factor determining physicians’ acceptance of mobile technology. Perceived ease of use had a rather limited and weak effect as physicians gained more hands-on experience with the mobile system; such effects varied according to different levels of a physician’s PIIT and age. The effects of ease of use might be significant for predicting physicians’ current usage behaviour, rather than future intentions towards the system. The perceptions of the compatibility of the mobile system with everyday work practice influenced a physician’s adoption behaviour significantly; this effect varied with the different levels of a physician’s age as well as with the different degree of a physician’s FWS. Gender, hands-on experience of the system and a physician’s FWT had different effects on physicians’ perceptions of the mobile system. The key contributions of the thesis are theory extensions and the testing of the TAM in new contexts, i.e. the contexts of different culture, health-care setting, the professional user group and mobile technology. Two new concepts, individual differences in FWT and FWS, have also been introduced and empirically tested. They seem to be important factors influencing users’ adoption of mobile technology. The research contributes to understanding user adoption of mobile technology in organisations, especially to understanding the adoption of mobile technology by professional users. Keywords: User adoption, technology acceptance model, mobile technology, health care, mobile medical information system, physicians, perceived usefulness, perceived ease of use, compatibility, fragmentation of working time, fragmentation of working space, behaviour intentions

i

ii

ACKNOWLEDGEMENTS This work presents the results of my Ph.D studies at the Turku Centre for Computer Science (TUCS) and the Institute for Advanced Management Systems Research at Åbo Akademi University over the past four years. Thanks are due to many people who have supported me. First, my sincere thanks go to my supervisors, Professor Pirkko Walden and Professor Christer Carlsson, for their endless support and constructive suggestions. I would like to thank Professor Christer Carlsson also for having recommended me to join the TUCS graduate school in 2001, and for his strong leadership, which has made the Institute a great place to do research in. I am also grateful to Professor Barbro Back for her commentaries and encouragement, particularly during the postgraduate seminars. The administrative staff, Stina, Leena, Pia and Sirpa, has always been kind and helpful during the dissertation process. I owe special thanks to Professor Per E. Pedersen from the Agder University College, Norway and Professor Virpi K. Tuunainen from the Helsinki School of Economics and Business Administration, Finland, who kindly functioned as my reviewers and finished their excellent work in a month. Their comments have helped me considerably improve the quality of the thesis. I would also like to thank Professor Harry Bouwman from the TU Delft, the Netherlands for accepting the invitation to function as my faculty opponent at the public examination. I am extremely grateful to Dr. Markku Kallio, Dr. Pekka Mustonen and Dr. Matti Seppänen from Duodecim – the Finnish Medical Society. They have given me endless help in collecting the empirical materials and also shared their experience in doing scientific research during the Duodecim project. I would also like to thank Juhani Toimela from Pfizer Finland Ltd., for sharing his experience and for the generous financial support to the conducting of my empirical studies. I also want to thank all the physicians, who took part in the Duodecim project, for their involvement and commitment. I sincerely thank Professor Leif Nordberg from the Åbo Akademi University for his constructive suggestions for statistical methods for the data analysis. I would like to thank Professor Pertti Järvinen from the University of Tampere for his comments on an earlier version of chapter three, i.e. the research approach. I would also like to thank Christopher Grapes for correcting the language of all my publications, the doctoral dissertation included. Thanks also go to Dr. Erkki Patokorpi for his help in improving the language of the thesis. I would like to express my gratitude to the entire staff of TUCS for giving me financial and administrative support during my studies. Three other organisations have also supported this work financially. I gratefully acknowledge the support of the Nokia Foundation, Pfizer Finland and Stiftelsens för Åbo Akademi forskningsinstitut . I am very grateful to the members of the mobile commerce laboratory, Chihab BenMoussa, Joanna Carlsson, Ville Harkke, Vaida Kadyte, Pär Landor, Dr. Shuhua Liu, Ruggerio Rossi de Mio, Anna Sell and Dr. Franck Tétard. In particular, Ville invited me to join the Duodecim project, which gave me a unique opportunity to conduct my empirical studies. Ville has always supported me and iii

been a recurring co-author. Franck has always kindly shared his knowledge and experience with me in various discussions. I would also like to thank my colleague Dr. Tomas Eklund for his suggestions of finishing the thesis. My Chinese friends deserve my gratitude for making my days in Åbo so unforgettable. He Bo, Dr. He Tao, Dr. Niu Li, Wang Zhao, Liu Yang, Lu Hao, Dr. Zhu Jing, Dr. Gu Ningyu, Zhang Di, Duanmu Jie, Yu Xiang, Wang Shaoxia, Su Pingping, Yan De, Xu Chunlin, Tan Jiaxi, Wei Di, Dr. Zhang Peng, Yan Zheng, Yan Ping and many others, have made my life full of fun and laughter. In particular, He Bo has always taken care of me. He Tao and Zhang Di organised the most incredible birthday party of my life. Liu Yang has always been with me to go through hard times. Gu Ningyu has checked all the references of the dissertation. I would also like to thank Professor Luo Guifen for her encouragement and care. I am very grateful for the care and help that Marita Siika and Ilkka Heikkinen have given to me. They have treated me as their daughter and friend. Without their love and encouragement, I could not have coped with all the difficulties I met during my studies in Finland. Last but not least, my warmest thanks go to my family – my parents, my brother and my sister-in-law. My parents have always been there to support all my decisions and they have been very proud of my achievements. Thank you, Mum and Dad, for your love, passion, and dedication.

Shengnan Han Åbo, March 28, 2005

iv

TABLE OF CONTENTS PART ONE: RESEARCH SUMMARY Chapter 1 Introduction............................................................................................ 1 1.1 1.2 1.3 1.4

Motivation and Research Objective ........................................................ 1 Research Questions ................................................................................. 3 Structure of the Dissertation .................................................................... 4 Contributions and Publications................................................................ 5

Chapter 2 Background ............................................................................................ 9 2.1 Mobile Commerce: Fundamental Definitions and Concepts................... 9 2.2 Mobile Services for Physicians ............................................................. 12 2.3 The Duodecim Project ........................................................................... 13 2.3.1 The Roadmaps of Mobilising Medical Knowledge for Physicians in Finland...................................................................................... 14 2.3.2 The Mobile System and the Target User Group ........................... 14 Chapter 3 Research Approach.............................................................................. 17 3.1 Diversity in Research on Information Systems ..................................... 17 3.1.1 The Information Systems Discipline ............................................. 17 3.1.2 Philosophy of Science for Information Systems Research ............ 18 3.1.3 Research Methods in Information Systems Science...................... 19 3.1.4 Discussion..................................................................................... 20 3.2 Quantitative Positivist Research ............................................................ 21 3.2.1 Rigour in Quantitative Positivist Research................................... 21 3.2.2 Relevance in Quantitative Positivist Research ............................. 24 Chapter 4 Technology Acceptance Model (TAM) .............................................. 27 4.1 4.2 4.3 4.3.1 4.3.2 4.3.3 4.4 4.4.1 4.4.2 4.5 4.5.1 4.5.2 4.5.3 4.6 4.7 4.8 4.9 4.10

The Origin ............................................................................................. 28 Model Adoption..................................................................................... 29 Model Extensions .................................................................................. 33 Dimensions of Usage .................................................................... 33 Moderators.................................................................................... 34 Extensions of PU and EU ............................................................. 35 Model Validation................................................................................... 37 Instrument Validation ................................................................... 37 Validation of Casual Relationships .............................................. 38 Model Limitations ................................................................................. 41 Cultural Dimension of the TAM.................................................... 41 Applicability and Generalisability Issue....................................... 42 Measurement of Usage ................................................................. 42 Comparison with Other Theories .......................................................... 43 The Unified Theory of Acceptance and Use of Technology ................. 44 Adoption Research for Mobile Technology .......................................... 45 Empirical Studies of Physicians’ Acceptance of Technology ............... 47 Discussions and Conclusion.............................................................. 47 v

Chapter 5 Time, Space and Mobile Technology ................................................. 51 5.1 5.1.1 5.2 5.3 5.4

Introduction ........................................................................................... 51 Mobility and the Use of Mobile Technology: A Brief Review....... 52 Temporal Working Behaviour: Fragmentation of Working Time ........ 55 Spatial Working Behaviour: Fragmentation of Working Space ........... 55 FWT, FWS and Physicians Work.......................................................... 56

Chapter 6 Research Models .................................................................................. 59 6.1 6.2 6.3

Research Questions Revisited ............................................................... 59 Research Models ................................................................................... 62 Research Design and Data Analysis...................................................... 64

Chapter 7 Physicians’ Adoption of Mobile Technology ..................................... 69 7.1 Insights from a Practice Perspective...................................................... 69 7.1.1 Insights from Two Telephone Interviews ...................................... 69 7.1.2 Insights from the Survey ............................................................... 71 7.1.3 Summary: A Practice Perspective ................................................ 75 7.2 Insights from a Research Perspective .................................................... 75 7.2.1 Testing Research Model 1............................................................. 75 7.2.1.1 The data................................................................................ 75 7.2.1.2 The research hypotheses and the results .............................. 76 7.2.1.3 Discussion ............................................................................ 78 7.2.2 Testing Research Model 2............................................................. 79 7.2.2.1 The data................................................................................ 79 7.2.2.2 The Research hypotheses and the results ............................. 80 7.2.2.3 Discussion ............................................................................ 82 7.2.3 Summary: A Research Perspective ............................................... 84 7.3 Concluding Remarks ............................................................................. 85 Chapter 8 Conclusions........................................................................................... 87 8.1 8.2 8.3

Contributions to and Implications for Research .................................... 87 Contributions to and Implications for Practice...................................... 88 Limitations and Future Research ........................................................... 90

References............................................................................................................... 91 Appendices............................................................................................................ 105 PART TWO: ORIGINAL RESEARCH PAPERS ........................................... 113

vi

LIST OF FIGURES AND TABLES Figure 1 M-commerce: Key Elements of Products and Services ........................... 11 Figure 2 The Mobile Medical Information System................................................. 15 Figure 3 Technology Acceptance Model (TAM).................................................... 29 Figure 4 Unified Theory of Acceptance and Use of Technology (UTAUT) .......... 45 Figure 5 Research Model 1 ..................................................................................... 63 Figure 6 Research Model 2 ..................................................................................... 63 Table 1 The Primary Research Questions................................................................. 3 Table 2 Overview of the Dissertation ....................................................................... 5 Table 3 Definitions of Mobile Commerce ................................................................ 9 Table 4 Validation Guidelines Adopted in the Dissertation ................................... 23 Table 5 Literature of the TAM................................................................................ 30 Table 6 Validation of the Relationships-- the TAM ............................................... 39 Table 7 The Research Questions Revisited............................................................. 61

vii

LIST OF ABBREVIATIONS IS – Information Systems IT – Information Technology QPR – Quantitative Positivist Research TAM – Technology Acceptance Model UTAUT – Unified Theory of Acceptance and Use of Technology FWT – Fragmentation of Working Time FWS – Fragmentation of Working Space PU – Perceived Usefulness EU – Perceived Ease of Use SI (SN) – Social Influence (Social Norms) COMP – Compatibility PIIT – Personal Innovativeness in the domain of Information Technology BI – Behavioural Intention TRA – Theory of Reasoned Action TPB – Theory of Planned Behaviour PCI – Perceived Characteristics of Innovating DIT – Diffusion of Innovations Theory GP – General Practitioner SP – Specialist EBMG – Evidence-Based Medical Guidelines ICD-10 – International Diagnosis Code guide, version 10 XML – eXtensible Mark-up Language SMS – Short Message Service MMS – Multimedia Messaging Service PDA – Personal Digital Assistant GSM – Global System for Mobile Communications GPRS – General Packet Radio Services UMTS – Universal Mobile Telecommunications Systems

viii

LIST OF ORIGINAL PUBLICATIONS

Paper 1 Han, S., Harkke, V., Landor, P., and Rossi de Mio, R. (2002) ‘A foresight framework for understanding the future of the mobile commerce’, Journal of System and Information Technology, December 2002. Paper 2 Han, S., Harkke, V., Mustonen, P., Seppänen, M., and Kallio, M. (2005) ‘Understanding physician acceptance of mobile technology: insights from two telephone interviews in Finland’, Int. J. Electronic Healthcare, Vol.1, No.4, pp.380-395. Paper 3 Han, S., Harkke, V., Mustonen, P., Seppänen, M., and Kallio, M. (2004) ‘Mobilizing medical information and knowledge: some insights from a survey’, In Proceedings of the 12th European Conference on Information Systems(ECIS), Turku, Finland, 13-16 June, 2004. Paper 4 Han, S., Mustonen, P., Seppänen, M., and Kallio, M. (200x) ‘Physicians’ acceptance of mobile communication technology: an exploratory study’, Int. J. Mobile Communications (to appear). Paper 5 Han, S., Mustonen, P., Seppänen, M., and Kallio, M. (2004) ‘Physicians’ behaviour intentions regarding the use of mobile technology: an exploratory study’, In Proceedings of the 8th Pacific Asia Conference on Information Systems(PACIS), Shanghai, China, 8-11 July, 2004. Paper 6 Han, S., Mustonen, P., Seppänen, M., and Kallio, M. (2005) ‘Physicians’ acceptance of mobile technology: investigating the effects of fragmentation of working time and working space’, Submitted to Electronic Commerce Research and Applications (in review process).

ix

x

PART ONE: RESEARCH SUMMARY

Chapter 1 Introduction “… the more people adopt a particular technology, the more it improves, and the more incentive there is for further adoption.” W.B. Arthur (1990)

1.1

Motivation and Research Objective

Mobile technology is transforming communication between human beings. It has a profound effect on societal change around the world. Studies by the Gartner Group (2000) showed that data-enabled wireless devices would exceed the number of Internet-connected PCs by the year 2003. The prediction is true. The world is crowded with mobile phone users, from the poorest villages in Africa and Asia to the most industrialised cities in Europe and North America. Business is changing profoundly as well. Mobile commerce, or mcommerce, is already creating many innovations. Recent years have seen the adoption of mobile technology as a potential aid for business activities. It is clear that although mobile technology has not been able to fulfil all its expectations to change the basic nature of business, it is a profitable innovation that opens up new business opportunities and offers added value for both the customer and the company (Barnes, 2002; Tarasewich et al., 2002). Yet, in order for mobile technology to bring value, it must be accepted and used by users or customers. It is important to study the adoption of mobile technology. My choice of adoption studies is based on the following rationales. First, in the mobile technology context, it is the users or customers who know what kind of value they need. The m-commerce products and services are bought on this basis (Carlsson and Walden, 2002a). Satisfying the wants or needs of customers is a key factor for success (Kalakota and Robinson, 2001). Mobile technology improves as more people adopt it and gain experience of it. Second, if the users or customers do not accept the products or services, the value provided by the company will not be realised. Since, in modern high-technology economies, products and services are knowledge-based, they require a large initial investment in research, development and tooling (cf. Shapiro and Varian, 1998). If the producers and suppliers fail to meet the value expectations of the customers, the initial investment will be wasted and the companies will lose their core competence. Third, adoption is a decision process (Rogers, 1995), and this process is influenced by many external factors (Ajzen and Fishbein, 1980; Davis et al., 1989). The greater the understanding of these factors, the more producers/suppliers can effectively manage interventions and build relationships with customers in order to improve product design, to promote implementation and to launch new products at the right moment and in the right place. Fourth, if producers/suppliers are able to promote the adoption of mcommerce products and services by customers, they will increase their returns in the long run. To conclude, the business value of mobile technology exists - and it is 1

huge - if companies understand their “target users” intimately and develop mobile products and services that recognise the role that “mobility” plays in users’ working and daily lives. We have to be aware that it is the users and their use, not advanced mobile technology, that will drive its growth to a new level (Jarvenpaa et al., 2003). As Peter F. Drucker (1954: p.37) wrote, “there is only one valid definition of business purpose: to create a customer. It is the customer who determines what a business is”. Hitherto, users’ perceptions of and intentions to adopt information systems (IS) and the rate of diffusion and penetration of technology within and across organisations have been two important foci of IS research (e.g. Straub et al., 1995; Taylor and Todd, 1995a). They are understood to represent the essential aspect, property or value of information technology (Orlikowski and Iacono, 2001). It is generally accepted that the use of information systems at work could increase employees’ productivity and improve both the individual’s and the organisation’s performance. System usage is an important outcome of measuring IS success (DeLone and McLean, 1992 and 2003). In general, this dissertation seeks to delineate the emerging research field that is called adoption research of mobile technology from the individual user’s or consumer’s perspective (Methlie and Pedersen, 2001; Pedersen, 2002; Pedersen and Ling, 2003). It is a research field in the domain of the management of IS and addresses the effects of important factors and individual beliefs influencing the decision-making process regarding the use of mobile technology. In particular, this research has a special motive for investigating physicians’ behaviour regarding using mobile technology in their work practice. In the last three decades, information technology (IT) has slowly become an integral part of health-care practice, management and processes (Ammenwerth et al., 2004). As investment in IT by the health-care industry continues to grow rapidly, efficient use and mass adoption of IT by physicians has emerged as a critical technology implementation and management issue (Chiasson and Davidson, 2004; Wilson and Lankton, 2004). Regardless of its potential technical advantages, an unused or under-used technology cannot create added value for health-care improvement, however. Recent years have seen mobile technology starting to be adopted by the health-care industry. Goldberg and Wickramasinghe (2003) have argued strongly that mobile e-health services offer a “panacea” for health-care problems in the 21st century. Their views direct attention to new phenomena, e.g. the design and use of mobile technology in health-care settings (see also: Wickramasinghe and Goldberg, 2004). A number of companies are extending their Internet services for physicians to use with Personal Digital Assistants (PDAs) or other mobile terminals. The use of PDAs among doctors is rising, having reached 235,400 by 2004 in the U.S. (Mobile health data, 2004). In Europe the nations with the highest percentages of general practitioners who use PDAs in their practices are: the Netherlands (31%); the United Kingdom (18%); Spain (17%); France (11%); and Germany (10%) (Harris interactive, 2002). A recent survey of paediatricians who use PDAs in their practice claims that PDAs improve health-care quality. The improvement results from handheld access to drug and clinical references at the point of care (Mobile 2

pipeline, 2004). In Finland the PDA has now become the tool of choice for medical doctors; smart phones, such as the Nokia Communicator, have a much wider user base. The primary objective of the research is to achieve these general and special aims, i.e. to contribute knowledge about physicians’ adoption of mobile technology, and their intentions to use this technology in the health-care setting in Finland. 1.2

Research Questions

The technology acceptance model (TAM) (Davis, 1989; Davis et al., 1989) is one of the robust models that IS researchers have adapted intensively to predict and explain a user’s adoption of various computer technologies in vastly different working settings. When mobile technology is used in health-care context, the applicability and adaptability of the TAM should be tested and modified because: (i) individuals use mobile technology “ on the move”, free from physical office constraints and use it “anywhere”; (ii) individuals use mobile technology “anytime”, free from linear working time. Time is becoming more and more socially constructed and negotiated, which blurs the boundary between working time and leisure time; (iii) Individuals use mobile technology in their “mobile” work, which differs vastly from their “static” office work; (iv) in the mobile context, perceived usefulness (PU) may not have the same significant prediction and explanation power to explore individuals’ use of mobile technology. Other factors may emerge as more important determinants of an individual’s decision to use mobile technology, e.g. individual differences in fragmentation of working time (FWT) (Tétard, 2002) and fragmentation of working space (FWS) (e.g. Kristoffersen and Ljungberg, 1998; Luff and Heath, 1998; Tengblad, 2002), social influence, and compatibility etc. According to Chiasson and Davidson (2004), (v) the concepts and theoretical assumptions of IS theories need to be reshaped and re-examined to fit the health care industry context. Since my special focus is on physicians’ behaviour in the health-care setting in Finland, the primary research questions are (Table 1): Table 1 The Primary Research Questions

Questions1: Why would physicians adopt mobile technology? Question 2: What factors influence physicians’ adoption of mobile technology? Question 3: Are these factors explaining and predicting physicians’ adoption of mobile technology consistent with the TAM? Question 4: Could the FWS, FWT factors enhance our knowledge of the adoption of mobile technology? By answering these questions, I aim (i) to explore the important factors that influence users’ adoption of mobile technology in health-care settings. The results might provide useful information for managers needing to assess the likelihood of success when introducing mobile technology and help them understand the drivers 3

of acceptance in order to proactively design interventions (e.g. motivation policy, training programmes) targeted at populations of users that may be less inclined to adopt and use new systems; (ii) especially to examine the effect of individual differences in FWT and FWS on users’ (physicians’) adoption of mobile technology; (iii) to generalise my findings and to prescribe some impacts for the development of mobile technology. In order to fulfil the research objective and find answers to the research questions, I conducted empirical studies in close research cooperation with Duodecim Medical Publications Ltd.,1 owned by Duodecim, the Finnish Medical Society, from June 2003 to the end of 2004. There are two kinds of data sets. The first one was collected by Duodecim with support from Pfizer Finland. The methods used were telephone and mail interviews 2 . The second data set was collected by me with support from Duodecim. It was gathered by a mail interview and a web survey3. 1.3

Structure of the Dissertation

The dissertation consists of two parts. Part one is a summary of my doctoral research process, which I have conducted over the past four years. Part two is a collection of six original articles. Part one is divided into eight chapters. This chapter begins with an explanation of the motive, research objective and primary research questions. It is followed by an outline of the structure of the dissertation. Chapter 2 begins with the definitions of key concepts that have been used in the dissertation, i.e. m-commerce, m-commerce products and services. It also provides reasons for providing mobile services for physicians. At the end, the study context and the Duodecim project are described in detail. It is partly based on the results of papers 1, 2 and 3. Chapter 3 provides a methodological basis for the research project. Discussions of quantitative positivist research (QPR) in IS are presented. It is an issue that is not offered in the original papers. Chapter 4 presents a thorough literature review of the TAM and the unified theory of acceptance and use of technology (UTAUT). These are the underling theories of the dissertation. This chapter gives preliminary answers to the research questions from a theoretical perspective. Papers 4, 5 and 6 have reviewed these theories briefly. 1

Duodecim Medical Publications Ltd. is the Scandinavia’s largest medical publisher. It specialises in medical textbooks, manuals, dictionaries, and electronic publishing. Duodecim’s Internet portal is used in all Finnish public hospitals and more than 90% of primary care health centres. To date, Duodecim has published more than 100 titles, covering most medical specialities. The company is owned by Duodecim, the Finnish Medical Society. Founded 120 years ago, Duodecim has grown from an initial membership of 12 medical students, hence the Society’s name, to a nationwide organisation with over 17,200 members, or more than 80% of Finnish medical doctors and students. The Society publishes the most important scientific medical journal in Finnish and organises extensive educational programmes, as well as the five largest annual medical conventions in Finland. The Society also promotes continuous medical education and supports research through its own research foundation. (www.duodecim.fi) 2 cf. Tull and Hawkins (1987) 3 http://www.webropol.com

4

Chapter 5 introduces emerging important factors in the mobile technology context that might have potential effects on user acceptance of mobile technology, e.g. individual differences in FWT and FWS. Papers 5 and 6 have empirically studied these effects in an exploratory manner. Chapter 6 revisits the research questions presented in chapter 1, and proposes research models 1 & 2 as partial answers. The research design and data analysis adopted to answer research questions and to test research models are presented. Papers 4 and 6 have contributed insights to this chapter. Chapter 7 summarises the key findings from the original papers (except paper 1). The chapter is structured in accordance with different data sets, meaning that the results from the first data set are discussed separately with those from the second set. This chapter gives answers to the research questions revisited in chapter 6. Chapter 8 presents the main conclusions to be drawn from the research. The main contributions to IS research and practice are emphasised, the limitations of the study are pointed out, and some avenues for future research are also suggested. An overview of the dissertation is shown in Table 2. Table 2 Overview of the Dissertation

Topic Part I Chapter 1 Introduction Background: key concepts, mobile services for Chapter 2 physicians and the Duodecim project Chapter 3 Research Approach Chapter 4 Technology Acceptance Model (TAM) Chapter 5 Time, Space and Mobile Technology Chapter 6 Research Models Chapter 7 Physicians’ Adoption of Mobile Technology Chapter 8 Conclusions 1.4

Part II Paper 1 Papers 1, 2, 3 Papers 4,5,6 Papers 5,6 Papers 4, 6 Papers 2,3,4,5,6

Contributions and Publications

The key contributions of the thesis are theory extensions and the testing of the TAM in new contexts, i.e. the contexts of different culture, health-care settings, the professional user group and mobile technology. Two new concepts, individual differences in FWT and FWS, have also been introduced and empirically tested. They seem to be important factors influencing users’ adoption of mobile technology. The research contributes to understanding user adoption of mobile technology in organisations, especially to understanding the adoption of mobile technology by professional users. The TAM is a parsimonious model that has been applied in several hundred studies of user acceptance of different information technology artefacts, applications and services. In the literature, there have been a number of efforts to extend and test the TAM in the health-care settings (e.g. Hu et al., 1999; Chau and Hu, 2002a and 2002b), as well as in the mobile technology context (e.g. Junglas and Watson, 2003; Pedersen et al., 2003). However, little research in IS has been 5

done to examine the TAM for explaining and predicting physicians’ acceptance of mobile technology in health-care settings. Such a research is conducted and reported here. Two research models were proposed and empirically examined, using data collected in the Duodecim project, which involved 800 physicians who were working in the Finnish health-care sector in a pilot trial of a mobile medical information system from June of 2003 till the end of 2004. This system is operated with the Nokia Communicator 9210. The results suggested that the two proposed research models could adequately explain physicians’ intentions to use mobile technology. Based on the results achieved, a number of implications for IS research and the management of mobile technology for promoting a mass adoption are presented. Thus, the findings provide guidance to health care organisations that are implementing mobile technology. During the course of the incremental research, the results have been reported in a number of scientific papers. Paper 1, “A foresight framework for understanding the future of the mobile commerce”, introduces the key concepts and definitions which are relevant in the mobile commerce context. Using data gathered from published forecasts and surveys, with the help of an information-gathering software-agent, four industry foresight scenarios are presented with a focus on the adoption of mobile commerce and the general state of the economy. This paper serves as the point of departure for the dissertation. In the paper, we discuss the importance of adoption and diffusion processes for understanding the future of mobile commerce. The fundamental definitions of mobile commerce, mobile commerce products and services, and killer bouquets are the core part of chapter 2. Paper 2, “Understanding physician acceptance of mobile technology: insights from two telephone interviews in Finland”, looks into the information needs of physicians and the mobilisation of medical knowledge. With the TAM as a theoretical point of departure, the paper reports on descriptive data gathered by telephone interviews of 42 physicians both in June and October, 2003. The interviewed physicians had positive perceptions of the mobile system, and started to use it in their work frequently. The study provides practical knowledge of physicians’ behaviour regarding mobile technology. Paper 3, “Mobilizing medical information and knowledge: some insights from a survey”, discusses the results from a large empirical survey which was conducted from April to September, 2003. It is an empirical enhancement of the results from paper 2 and enriches our practical knowledge concerning the physicians’ information needs and usage of the mobile medical information system. Paper 4, “Physicians’ acceptance of mobile communication technology: an exploratory study”, presents a research model, i.e. research model 1 of the dissertation, on physicians’ intentions to use mobile technology based on the TAM, UTAUT and personal innovativeness in the domain of IT (PIIT). The model is tested empirically with data collected from November 2003 to January 2004. The key contribution is in extending the TAM’s theoretical validity and empirical applicability with moderation effects of PIIT and age in the context of health care and mobile technology. 6

Paper 5, “Physicians’ behaviour intentions regarding the use of mobile technology: an exploratory study”, presents a research model, based on the TAM, UTAUT and on the effects of individual FWT and FWS influencing the physicians’ willingness to accept mobile technology. The empirical data used in the exploratory study is the same as in Paper 4. The model proposed here was the first attempt to explore possible relationships between individual FWT and FWS and physicians’ perceptions and intentions toward mobile technology. This was served as a basis for research model 2, which was further worked on with new empirical material and reported in Paper 6. Paper 6, “Physicians’ acceptance of mobile technology: investigating the effects of fragmentation of working time and working space’, proposes a research model, i.e. research model 2 of the dissertation by improving and revising the research done in Paper 5. New data were collected by a web survey to empirically examine the model. The key contribution of the paper is the inclusion of more empirical data and theoretical enhancements, focusing on the relationship between physicians’ FWT and FWS, and their adoption of a particular mobile information system.

7

8

Chapter 2 Background In this chapter, fundamental definitions and concepts related to mobile commerce are introduced. Physicians’ need for using mobile technology is discussed. The details of the Duodecim project are presented at the end. 2.1

Mobile Commerce: Fundamental Definitions and Concepts

Mobile commerce (m-commerce) has hit the business world in the recent years. Many experts suggest that the first decade of the 21st century will be the decade of mobile computing and m-commerce (Urbaczewski et al., 2003). Regardless of the vast publicity and the widespread familiarity of this term, there is no generally accepted definition of mobile commerce among scholars or business practitioners. Several different definitions have been proposed. Some examples are presented below (Table 3): Table 3 Definitions of Mobile Commerce

Müller-Versee (2000): m-commerce as a vast area of activity comprised of transactions with monetary value conducted via wireless network. Skiba, Johnson and Dillon (2000): m-commerce as transactions using a wireless device either in word or speech that result in the transfer of value in exchange for information, services, or goods. Mennecke and Strader (2002): m-commerce, namely, e-commerce activities carried out via a mobile device, such as a cell phone or PDA. Sadeh (2002): m-commerce involves an emerging set of applications and services people can access from their web-based mobile device. Varshney and Vetter (2002): …many new e-commerce applications will be possible and significantly benefit from emerging wireless and mobile network. We term these applications “wireless e-commerce” or “mobile commerce”. Mathew, Sarker and Varshney (2004): m-commerce refers to the emerging arena within which commercial transactions are made possible using handheld mobile devices that are connected by wireless networks. In this dissertation m-commerce is defined as the extension of electronic commerce (e-commerce) from wired to wireless computers and telecommunications, and from fixed locations to any time, anywhere and anyone (Keen and Mackintosh, 2001), i.e. the use of mobile technologies and devices to provide, sell and buy convenient, personalised, and location-based services. The consumers of mobile commerce are understood to be any customers, employees and/or business partners; thus, B2B (business-to-business), B2C (business-toconsumer), B2E (business-to-employee), and P2P (peer-to-peer) are becoming the contexts of mobile commerce. From the supply side, numerous companies are needed in the mobile value chain to produce mobile services, ranging from technology providers and network operators to content providers, etc. (Barnes, 2002). 9

Carlsson and Walden (2001; 2002a and 2002b) provide a conceptual framework for m-commerce products and services from three perspectives: the customer, the producer and management. M-commerce products and services viewed from different perspectives have different distinguishing elements. These embedded elements actually describe what the m-commerce products and services will be. From the customer perspective, they are durable technological products and technology-based services where repeat purchases are triggered by an increase in user-perceived functionality. From the producer perspective, they are new concept products that will boost companies’ profits or even create a sustainable competitive advantage. From the management perspective, they are new information systems for decision-making or business models that need a new strategy to push companies ahead of their competitors in the future. Seen from the perspective of the customer the necessary distinguishing elements of m-commerce products and services are: 1. Flexibility: m-commerce products and services should be available anywhere, at any time and anyhow. 2. Value-adding: m-commerce products and services should improve productivity, they should be adaptable to localisation and they should be sensitive to customer personalisation. 3. A mobile technology basis: m-commerce products and services should use innovative and distinguishing features of mobile technology to enhance the quality of life (e.g. messaging, entertainment, education, information, privacy, etc.). Seen from the perspective of the producer the necessary distinguishing elements are: 4. Modularity: m-commerce products and services should be built from a core of generic product and service modules that can be combined to form context-adapted products and services; this should support the flexibility element. 5. Layers: m-commerce products and services could be built in layers to add attributes and characteristics that can be adapted to (i) customer personalisation, (ii) localisation, (iii) brand profiles, (iv) privacy etc.; this should support the value-adding element. 6. Bundling: m-commerce products and services should be built through a bundling of modular products and services, which would be a way to make use of the mobile technology basis. Bundling can be done through modules and layers, but can also be mobile technology-based. Seen from the perspective of the management the necessary distinguishing elements are: 7. Value/cost ratios: m-commerce products and services should show good or very good value for cost compared with similar products and services; this should form the basis for pricing strategies, and cost and revenue models. 8. Production, logistics, marketing and advertising: m-commerce products and services should have innovative features compared with similar products and services; this may be a function of the possibilities offered by mobile technology. 10

9. Business model: m-commerce products and services should use innovative and distinguishing features of mobile technology to support new business models. All these elements are shown in an intuitive graphical description (Fig.1). Modules Layers

LOCAL

LOCAL PERS

PROD 1 PERS – Personalisation LOCAL-Localisation UBI-Ubiquity TIME-Timeliness CON-Convenience PRICE-Pricing

SERV 1

PERS

UBI

CON

TIME

PROD 2

PRICE

UBI

CON

TIME PRICE

Bundle LOCAL PERS

LOCAL UBI

CON

PERS

TIME

SERV 2

PRICE

PRODUCER PERSPECTIVE: Modularity, Layers, Bundling

UBI

CON

TIME PRICE

(Carlsson and Walden, 2001)

Figure 1 M-commerce: Key Elements of Products and Services

In Fig.1, products and services are shown as “thick sets”, which are built with multiple layers (cf. the producer perspective), and each layer may be defined by multiple attributes, which may be the same or different for each layer. The products and services are described with multiple attributes that represent, for example, the key factors of success, the distinguishing elements (from user, producer or management perspectives), or some other features that are essential for the design of good combinations of m-commerce products and services. The attributes can be specific for modules and/or layers, and they can be defined for specific products and services, or be specific for bundles of products and services. It appears that with these simple elements a considerable variety of m-commerce product and service alternatives can be described. It is unlikely that there will be any mind-blowing “killer application” on the mobile markets during the years to come. The killer applications are probably more country-, culture- and context-bound (at least that is the impression one gets when looking at different mobile behaviour in Japan, Finland, Southern Europe and the US). This has been evident in discussion of m-commerce products and services. There have been various types of combinations: Killer Cocktail, a mix in which the components cannot be distinguished (Nokia); Killer Pizza, a mix in which the components can be distinguished; Killer Bouquet, a set of components for which the aggregate is greater than the sum of its parts; Killer Soup, the more ingredients you put in, the better it gets – an operator will be needed to stir; Killer Fondue, as for the soup, but no operator is needed to stir. 11

Using these, no doubt rather stirring metaphors, the “killer bouquet” is a bundle of m-commerce products and services (cf. Fig.1). With an understanding of the key features and success factors it appears that the core of the m-commerce products and services is, (i) to develop value-added content, (ii) to make them adaptable to localisation, (iii) to make them flexible to localisation, (iv) to make bundles of products and services ubiquitous and adaptable to moving customers, (v) to ensure timely delivery, and (vi) to build or enhance user freedom. The framework can lead to new insights by providing a systematic basis for investigating the whole range of products and services in the context of mobile commerce or mobile technology. 2.2

Mobile Services for Physicians

It is not surprising that mobile technology has penetrated the health-care industry. It has been adopted in particular to distribute medical knowledge and information. There are several possible reasons. Firstly, patient care in most environments is by its very nature a mobile experience, and a physician who can access information at the point of care and the point of need has a clear advantage over a colleague who is tied to a traditional desktop terminal. The use of mobile systems might substantially save time and effort as well as reduce errors that might be not only costly and inconvenient but even fatal. Secondly, medical knowledge is changing constantly. It is not easy for physicians to keep their knowledge and information up-to-date to help in their patient care and patient management efficiently on the one hand and to maintain the level of their professional competence on the other (Jousimaa, 2001). Chamliss and Conley (1996) argued that physicians often had unanswered critical questions during patient care. The available medical literature can provide answers to most of the questions, but to search for and find these answers was time-consuming and expensive. Therefore, it was important to provide more efficient ways to help physicians seek and retrieve medical information or knowledge. Through a systematic review of physicians’ information needs (Smith, 1996) and their information-seeking behaviour (Dawes and Sampson, 2003), it was found that printed desk references were the most common sources used when looking for information, but electronic information resources supported by computer systems had to be developed to meet the physicians’ information needs and help them reduce the burden of information overload. They also argued that the best information sources for physicians should provide relevant, valid material that can be accessed quickly and with minimum effort. Likewise, the sources enabled physicians to keep up-to-date and improve knowledge transfer. Finally, they pointed out the most important features that need to be taken into consideration when a new information tool was designed; these were that it should be electronic, portable, fast, easy to use, connected to a large valid database of medical knowledge and patient records and serve both patients and physicians. Verhoeven et al. (2000) also emphasised these features in their empirical study where they examined which literature retrieval method was most effective for general practitioners (GPs) from the year of 1994 to 1997. They concluded that using printed sources might be the most effective medical literature retrieval method, but 12

electronic sources seemed to be the most efficient. It is clear that physicians need to be encouraged to use electronic sources. Thirdly, from the medical publishers’ perspective, it is a great challenge for them to distribute knowledge to physicians in a timely manner through a fast, easy and convenient delivery channel. Detmer and Shortliffe (1997) argued that in order to speed up the diffusion of medical knowledge, information that was up-to-date and relevant to physicians’ information needs should be available in all work settings: office, clinic, hospital ward, library, and the home (p.101). They also strongly recommended a combination of content, information science methods, and technology to improve the diffusion. Referring to the key features yielded from the framework discussed in section 2.1, the core of a mobile service for physicians should have value-added contents, be flexible anywhere, at any time and anyhow, should add value to physicians’ work by improving patient care and patient management at the point of need and at the point of care, and should be based on mobile technology in order to enhance timely delivery and freedom. Existing mobile services for physicians available on the market range from simple medical dictionaries to sophisticated patient data systems capable of handling digital images and lab test results. A recent review by Fischer et al. (2003) has indicated that mobile services have become valuable in various fields of medicine. The systematic review summarised possible mobile services to access medical literature, electronic pharmacopoeias, patient tracking, medical education, research, business management, and eprescribing etc. The mobile device they focused on was the PDA. Here, I will investigate how physicians use a mobile medical information system operated in a Nokia Communicator in the Finnish health-care sector. The system is aimed to help physicians access the medical knowledge at the point of care and at the point of need. The research objectives are aimed to explore what kind of factors influence physicians adoption behaviour regarding mobile technology. No unused or under-used mobile service for physicians can add value to improve the quality of health care. Based on previous research, I intend to contribute knowledge of physicians’ adoption of mobile technology, and their intentions to use this technology in the health-care setting in Finland. 2.3

The Duodecim Project

The Duodecim project took the form of cooperation between Duodecim Medical Publications Ltd., and the Institute for Advanced Management Systems Research (IAMSR) at Åbo Akademi University. The cooperation started in June 2003 and continued till the end of 2004. The project aimed to investigate physicians’ adoption and usage behaviour regarding the newly developed mobile medical information system in a pilot trial. The insight and knowledge gained from the project would enhance our understanding of physicians’ acceptance of mobile technology.

13

2.3.1

The Roadmaps of Mobilising Medical Knowledge for Physicians in Finland

In Finland, the first computerised medical information database was launched in 1989 and disseminated on disks. It mainly contained 20 Finnish guidelines dealing with common and important primary health-care problems. In 1991, a CD-ROM was published with a guidelines database as well as additional databases, e.g. Finnish medical journals, laboratory databases, pictures. Throughout the 1990’s, Duodecim Medical Publications Ltd. put more effort into improving evidencebased medical guidelines (EBMG) to make them more comprehensive and reliable. With the development of Internet technology, an Internet-based version 4 was introduced in October 2000. Translation of these into English was completed in the year 2000 and into Swedish in 2001. Use of the computerised EBMG’s by Finnish physicians was found to be very encouraging. They could usually find the information they were looking for, and their searches were usually completed within 5 minutes (Jousimaa, 2001). EBMG has become an important source of information for Finnish physicians. Currently, EBMG as well as other databases, e.g. drug, diagnosis, etc. can be accessed through various channels, e.g. printed books (published annually), Intranets in health-care centres or hospitals, or the national Internet portal Terveysportti5, which also contains links to many domestic and international providers of health-care information. In 2002, a mobile version of those databases was developed ready for a pilot trial. 2.3.2

The Mobile System and the Target User Group

The mobile system (Fig.2) under investigation is a set of medical information and knowledge databases 6 (Duodecim call it a “mobile package”). It contains the EBMG (available both in English and Finnish) with Cochrane abstracts, the pharmacology database - Pharmaca Fennica with wireless update service for a complete drug (medicine) price list, the international diagnosis code guide (ICD10) in Finnish, the emergency guide issued by Meilahti hospital, a medical dictionary of over 57 000 terms and a comprehensive database over related addresses and contact information relevant to health care (pharmacies, hospitals, health centres, etc.). The content of the system is generated by an XML database. The system functions in most mobile devices operated by different systems, e.g. Symbian, Palm OS and Windows CE. The device most commonly used as a platform in Finland is the Nokia 9210 Communicator. The mobile medical system is delivered on a 128 MB (now on a 256MB) memory card and is self-installing, containing the search engine, user interface programs and core databases. Currently, the update is delivered in the form of physical memory cards, the users returning the older ones. In the near future, it will be able to update itself partly or completely through the GPRS or UMTS wireless networks. In the autumn of 2003, the new drug (medicine) price list was successfully updated through the GSM 4

www.ebm-guidelines.com www.terveysportti.fi 6 “Medical databases go mobile”, http://www.hightechfinland.com/2004/healthcarelife/duodecim.html [Accessed on 10 January, 2005]

5

14

network. Not only physicians involved in the trial, but also others not involved are now able to get medical news in SMS form. The contents of the system are continuously updated by exchanging the memory card. There were 800 physicians involved in the pilot trial. They had received Nokia 9210 Communicators equipped with the mobile medical information system free of charge thanks to support from Pfizer Finland Ltd. They were all medical practitioners holding different positions in health care. The choice of the targeted physicians was decided by Duodecim; the sampling was partly randomised; it included an equal number of general practitioners (GPs) and specialists (SPs), i.e. 400 per each group. The research project was finished at the end of 2004. However, Duodecim is continuing the pilot trial in co-operation with Pfizer Finland and with those physicians who are willing to buy the mobile medical information system operated on a Nokia Communicator.

Figure 2 The Mobile Medical Information System

15

16

Chapter 3 Research Approach In this chapter, I will discuss the methodological foundations for this research and demonstrate my own research approach. I will focus on the quantitative positivist research approach that has been adopted for the research project. 3.1

Diversity in Research on Information Systems

3.1.1

The Information Systems Discipline

Research in IS reflects considerable diversity in research practice and methods. From the conventional point of view, the IS discipline was an emerging field with fundamentally different schools of thought. Researchers argued that the IS discipline combined computer science and management to study the application of IT in organisations and society (cf. Keen, 1980) on the basis of very different perceptions of its core concerns, and with different approaches to investigation within the field (Checkland and Holwell, 1998). Furthermore, researchers have generally accepted the idea that the IS discipline had “reference disciplines”; it was a consumer of concepts at the end of an intellectual food chain (Vessy et al., 2002). Starting from this view, IS research in general, was labelled as “multidisciplinary” (e.g. Land, 1993), or “interdisciplinary” (Lee et al., 1999). A new revised standpoint has emerged, according to which IS is becoming a reference discipline “in a discourse with other reference disciplines” (Baskerville and Myers, 2002: p.7). Baskerville and Myers further pointed out that the opinion was not against IS being an applied science developed to serve the traditional audiences of IS scholars and practitioners. However, it called strongly for establishing a leading role of IS within “the larger community of scholars interested in the development, use, and impact of information technology and systems in broadly defined social and organizational settings” (idem, p.8). They argued that it was impossible for IS alone to discover the “truth” of this broad research domain, but by exerting a leadership role, IS could transform research knowledge and explain the broad value of research discoveries to scholars from other fields. Hirschheim and Klein (2003) further argued that in order to contribute to the field’s continued prosperity, a “reflective perspective” of IS research was needed. Recently, Lyytinen and King (2004) proposed a broader concept of “market of ideas” to build an academic legitimacy for IS research constructed on three drivers “the salience of the issues studied, the production of strong results and the maintenance of disciplinary plasticity”. They argued that the real centre in the IS field has been and will be “constituted through a market of ideas in which scholars (and practitioners) exchange their views regarding the design and management of information and associated technologies in organized human enterprise” (idem, p.236). Those strong arguments indicate that it is beneficial for IS research to embrace all relevant disciplines in a manner of plasticity. Discussions of the IS discipline also face the issue of whether IS needs a unifying paradigm in order to progress as a scientific discipline, or not (Moody and 17

Buist, 1999). A paradigm refers to the basic system of beliefs or worldview that guides the investigator, not only in the choice of methods, but also in ontologically and epistemologically fundamental ways (Guba and Lincoln, 1994). Benbasat and Weber (1996) recommended that in the IS discipline, it was a necessity to maintain a paradigm (one or more) and controlled diversity. A paradigm “will serve to provide coherence to the IS discipline and to characterize the phenomena that make it different from other disciplines” (p.397). Controlled diversity means to avoid the possible cost of diversity in IS research to enhance the establishment of “fundamentally dominating” IS paradigms. Benbasat and Zmud (2003) also defined the “core properties” of the IS discipline to propose arguments for the need for better governance over IS research, the methodological demands of the discipline and even permitted and proscribed topics. However, coordinating this diversity to build a dominating paradigm seems impossible in the IS field. The core property of the dissertation is the underlying dominant theory, the TAM that quite rigorously determines the path of my research. Thereby, I am able to control my research to avoid much diversity. I also embrace the “market of ideas” approach by integrating some contextual knowledge to enrich the results of the research. 3.1.2

Philosophy of Science for Information Systems Research

A fundamental source of diversity is the differing philosophies followed by researchers. Interpretive, positivist, critical and critical realist research can all be found in IS research (e.g. Mingers, 2004; Deetz, 1996). Interpretive research makes the epistemological assumption that reality is subject to multiple interpretations, since reality consists of subjective interpretations and cannot be studied objectively. Interpretive research aims at understanding and analysing subjective interpretations and their consequences; thus it seeks a relativistic, rather than shared, understanding of phenomena (Orlikowski and Baroudi, 1991: p.5). However, the understanding and analysis are mediated by the researcher. The process employs inductive logic or inductive reasoning. Findings are reported from his or her own subjective frame of reference (e.g. Lee, 1999; Lee et al., 1997; Walsham, 1993, 1995a and 1995b). Several different stands can be found in the literature, for example, ethnography (Harvey and Myers, 1995), ethnomethodology (Bhattacharjee and Paul, 2001) and phenomenology (Boland, 1985; Mingers, 2001b; Arnold, 2003). Orlikowski and Baroudi (1991) also classified critical research in IS. It assumes that social reality is historically constituted and that it is produced and reproduced by people. People’s ability to change their socio-economic situation is constrained by different social, cultural and political factors. The study attempts to reveal the historical, ideological and contradictory nature of existing social practice. By contrast, the epistemological assumption of positivist research is that reality is objective and observable; we can gain reliable knowledge of it. Researchers have to follow the scientific method in order to eliminate possible subjectivity from the research process (e.g. Benbasat and Weber, 1996; Hirschheim et al., 1996; Iivari et al., 1998; Westland, 2004). Positivist research generally attempts to test theory, in an aim to increase the predictive understanding of 18

phenomena (Orlikowski and Baroudi, 1991). To test the empirical validity of a theory, positivist researchers usually craft a “hypothetico-deductive” model and employ hypothetico-deductive logic to make conclusions. Several surveys of literature have demonstrated that most IS research has been underpinned by a positivist philosophy (Mingers, 2003; Nandhakumar and Jones, 1997; Walsham, 1995a). A recent survey by Chen and Hirschheim (2004) has also pointed out that it still dominated 81 per cent of published empirical IS research from 1991 to 2001. Mingers (2004) introduced a new philosophy for IS, critical realism. He discussed problems derived from interpretivism and positivism. The problems with interpretivism are twofold. One is that interpretive research is essentially a practical activity aimed at producing useful knowledge rather than understanding the true nature of the world. The other is that interpretive researchers find little correspondence with standard philosophical theories; the interpretation of findings is very subjectively oriented. The problems associated with positivism resulted from the impossibility of pure, unmediated observation of empirical “facts”; thus positivist research might not be able to provide an understanding of the truth. In order to overcome those problems, the philosophy of critical realism aims to “(i) re-establish a realist view of being in the ontological domain whilst accepting the relativism of knowledge as socially and historically conditioned in the epistemological domain; and (ii) argue for a critical naturalism in social science” (p.91). The author asserted that critical realism was important for IS because: “(i) it enables us to take a basically realist stance whilst accepting the major critiques of naïve realism; (ii) it addresses both natural and social science and thus encompasses the main domains of IS; and (iii) does potentially fit well with the reality of IS as an applied discipline” (idem, p.97). My research is largely dominated by the philosophy of positivism in IS research. In addition, I try to integrate the idea of “critical realism” by studying the context within which my study is conducted. This means that I try to increase the fit between the reality of the investigated phenomena and the positivistic observations I obtain from my empirical studies. 3.1.3

Research Methods in Information Systems Science

Research methods can be classified in different ways. However, there are two dominant groups of research methods in IS, quantitative and qualitative (Myers and Avison, 2002). Quantitative methods have originated from the natural science; later they have been applied in the social sciences through survey methods, laboratory experiments, formal methods (for example, econometrics) and numerical methods such as mathematical modelling. These methods and techniques are designed in such a way that numbers are assigned to represent values and levels of theoretical constructs and concepts. Also the interpretation of the numbers provides strong scientific evidence of how a phenomenon exists. The presence of quantities enables researchers to use statistical tools and packages to analyse observations. The emphasis on numerical analysis is also a necessary condition to meet positivist assumptions.

19

Qualitative methods 7 have developed in the social sciences to help researchers to investigate social and cultural phenomena. They are useful in understanding people and the social and cultural contexts within which they live. Qualitative methods include, for example, case-study research and ethnography. Qualitative data sources include observation and participant observation (fieldwork), interviews and questionnaires, documents and texts, and the researcher’s impressions and reactions. Depending on the underlying philosophical assumptions of a researcher, qualitative research could be positivist, interpretivist or critical. Qualitative positivist research indicates that the choice of a specific qualitative research method (such as the case-study method (Yin, 2002)) is underlined by the positivist philosophy of science (see review: Dubé and Paré, 2003). 3.1.4

Discussion

There are three different schools of thought to react to the aforementioned diversity in IS research, especially in terms of philosophical approach and methods. Imperialists8 are eager to build the dominance of one particular “IS identity”, either on philosophical assumption or on the necessity to create a strong discipline (Benbasat and Weber, 1996). Isolationists are willing to accept the thought of Burrell and Morgan (1979) that research should be developed separately because of the distinctive nature of various disciplines, philosophies of science and methods. In most cases, they are not commensurable. Recent years have seen the pluralist school of thought becoming popular. Pluralists either welcome diversity as it is, or believe that different methods could be more or less appropriate for a particular research question or situation, or recommend that research should work with diversity to triangulate phenomena (e.g. Van Maanen, 1995; Robey, 1996; Goles and Hirschheim, 2000; Mingers, 2001a). To summarise, the debate on diversity in IS research is open (cf. JAIS Vol.4, 2003 and CAIS Vol.12, 2003). It seems that diversity is a source of strength and vitality for the IS field, not a sign of weakness (Lyytinen and King, 2004; King and Lyytinen, 2004). Diversity might continue to characterise research efforts in IS. An extensive analysis of the debate is beyond the scope of this dissertation. I believe that the selection of a research approach (philosophy, methods) is incumbent on the individual IS researcher depending on the problems he or she aims to solve. Obviously, different research objectives require the use of different research approaches. As suggested by Moody and Buist (1999), the real question is not whether the research method is appropriate per se, but whether it is appropriate to answer the questions being asked. My research is underpinned by a positivist philosophy. In order to answer my research questions I have investigated the adoption behaviour of a large group of physicians, which means that I have adopted a quantitative research method.

7

Myers, M., http://www.qual.auckland.ac.nz/ According to Mingers (2004), the terms “ imperialist” and “isolationist” follow the analysis by Reed, M. (1985).

8

20

3.2

Quantitative Positivist Research

The acceptance and adoption of IT services has been one of the most popular IS research topics (e.g. Davis et al., 1989, Taylor and Todd, 1995a). Lyytinen and Yoo (2002) have further pointed out that in a nomadic information environment9 the adoption of nomadic (ubiquitous) information systems at the individual level is also one of the important research issues. My general aim is to find out what factors explain the adoption and use of mobile technology by individuals and how time, space and individual characteristics influence individual decision-making processes. The theory that underpins my research is the TAM. It is obviously an endeavour to explain what is. It is a typical task in positivist science. I adopted a quantitative, positivist research (QPR) approach as the underpinning philosophy and methodology to guide the empirical studies for my research. The rationale for this selection is the evidence that quantitative, positivist research is one of the most time-honoured ways of studying human behaviour and the use of technology. It has accumulated a huge amount of knowledge for understanding these specific socio-technical phenomena both for the IS research community and for business practices in different industry settings. According to Straub et al.10, QPR is “a set of methods and techniques that allow IS researchers to answer research questions about the interaction of humans and computers”. It consists of two cornerstones. The first one is the emphasis on quantitative data. The second is the emphasis on the positivist philosophy of science. QPR is “orthogonal to analytical (math) modelling, a scientific method that typically depends purely on mathematical derivations and assumptions for interpreting reality. This difference stresses that empirical data gathering or data exploration is part and parcel of QPR, while the positivist philosophy deals with problem solving and the testing of the theories derived to test these understandings”. 3.2.1

Rigour in Quantitative Positivist Research

Rigour refers to how data are generated, collected, evaluated and interpreted. The principles determining how rigorous research is conducted are essential for evaluating the quality of research for any scientific community. Rigour is achieved by applying existing foundations and methodologies appropriately (March and Smith, 1995). (i) Data collection techniques In my research approach, questionnaires were used to collect the empirical data, which is the most widely used technique for QPR (Palvia et al., 2004; Chen and Hirschheim, 2004).They were administered by a mail interview and later through a web-based survey. In order to understand the institutional characteristics of the 9

“a heterogeneous assemblage of interconnected technological and organizational elements, which enables the physical and social mobility of computing and communication services between organizational actors both within and across organizational borders” (Lyytinen and Yoo, 2002: p.378) 10 Available at: http://dstraub.cis.gsu.edu:88/quant/ [Accessed on 15 January, 2005]

21

health-care sector and the physicians’ opinions regarding the mobile medical system, I also analysed data that Duodecim and Pfizer Finland have collected. Those data as mentioned earlier were collected by means of telephone and mail interviews. (ii) Data analysis techniques In order to explain what are those factors influencing physicians’ behaviour regarding mobile technology, suitable analysis techniques have to be used rigorously to eliminate possible subjectivity on the part of the researcher. Regression models, e.g. multiple linear regression, logistic regression and MANOVA, are commonly used and generally accepted techniques in IS research. Regression models analyse one layer at a time to discover the relationships between independent and dependent constructs (variables). To be exact, they usually execute two unrelated analyses; they first examine how items load on the constructs via factor analysis, and then carry out a separate examination of the hypothesised paths. Therefore, real-world processes can only be demonstrated by simple correlation-based models. Regression models are considered to be reliable and suitable statistical techniques in IS research. I used the SPSS versions 11.0 and 12.0 to analyse my empirical data. The pre-assumptions of a regression analysis (e.g. linear or multiple regressions) have to be confirmed in the data analysis. Any violations of these assumptions would lead to invalid estimates and wrong interpretations of the research results. The assumptions are: (i) the relationship between a dependent variable and independent variables is linear; (ii) the independent variables are nonstochastic. In addition, no exact linear relationship exists between two or more independent variables; (iii) the error has an expected value of zero for all observations; (iv) the error term has constant variance for all observations; (v) errors corresponding to different observations are independent and therefore uncorrelated; (vi) the error term is normally distributed. Verifying those assumptions is an important step to diagnose the analysis. Remedies for possible violations are necessary in the analysis. By following the scientific method in data analysis, I am able to obtain valid and reliable results to answer the research questions. Consequently, potential inferences made from the study sample to the whole population are generally objective and valuable both for science and practice. The limitation of regression analysis is that it cannot test the measurement and the structural model at the same time. Therefore, it might provide less information about the extent to which a research model is supported by the data than when other techniques (e.g. LISREL or PLS; cf. Gefen et al., 2000) are used. (iii) Data evaluation and interpretation It is demanding to adopt “standard” and “systematic” validation guidelines to maintain QPR rigour in the IS field (e.g. Boudreau et al., 2001; Gefen et al., 2000; Straub et al., 2004). They argued that without solid validation of the research, the scientific basis of IS research and the profession will be damaged. Researchers who 22

do not follow these rules cannot claim, either, that the positivist methods selected are useful in the quest for scientific truth. Straub (1989) and Straub et al. (2004) pointed out the validity touchstones for IS empirical/positivist research. They asserted that instrumentation validation (i.e. content validity, construct validity and reliability) is both a prior and primary validation. Internal validity and statistical conclusion validity are the next two sequential steps. They offered an analysis of state-of-the-art of research validities and suggested specific heuristics for research practice. In the following, I will briefly survey these mandatory practices and some highly recommended guidelines from their studies with a focus on those that I adopted in my research (Table 4), which are generally featured by QPR. Table 4 Validation Guidelines Adopted in the Dissertation Validity Touchstones 1.Instrumentation Validity Content validity

Construct validity Discriminant validity Convergent validity

Reliability Internal consistency 2. Internal Validity

3.Statistical Conclusion Validity

Definitions

Techniques adopted

Constructs are likely real and reliable; the instrument is likely measuring the right content. An issue of representation. The degree to which the score or scale being used represents the concept about which generalisations are to be made. An issue of operationalisation or measurement between constructs. The measurement items posited to reflect the construct differ from those that are not believed to make up the construct. Measurement items reflect one “latent” construct which show significant and high correlations with one another in comparison with the convergence of those items relevant to other constructs. An issue of measurement within a construct; a statement about measurement accuracy. Inferences regarding cause-effect or causal relationship are approximately true and only relevant to the specific study in question. The degree to which we reach conclusions about relationship in our data is reasonable; the statistical inferences made are approximately true.

Literature review: consensus

high

degree

of

PCA: Eigenvalues>=1; loadings of at least 0.40; no cross-loading of items >0.40; Items do not load properly are dropped. PCA: Eigenvalues >=1; loadings of at least 0.40; items load on posited constructs; items which do not load properly are dropped. Cronbach’s α>0.60/0.70 and 24 months 7. Temporal behaviour of working My work schedule is often influenced by unexpected events during my working day.1 2 3 4 5 I do work in my free time (non-office time). 12345 I usually do many things at the same time. 12345 I usually achieve what is planned in my typical working day. 12345 I usually try to keep to a specific schedule at work. 12345 My working day is built of routines 12345 My work schedule often depends on other peoples’ schedule. 12345 8. Spatial behaviour of working I usually do work in my own office. 12345 For different reasons, I visit other places (e.g. reception, wards) near my office during my working day. 12345 I do work at home sometime. 12345 I regularly visit other health care locations (other hospitals or health care centres) within or outside the same municipality. 12345 I do work on a transportation vehicle (e.g. bus, train, plane, or ferry etc.) 12345 I take “house call” visits frequently. 12345

© IAMSR/ÅA & DUODECIM

9. Personal innovativeness Among my colleagues, I am usually the first to try out new information technology. 1 2 3 4 5 I like to experiment with new information technology. 12345 When I hear about new information technology I aim to try it. 12345 In general, I am hesitant to try out new information technology. 12345 10. Perceived usefulness I find the mobile package useful in my practice/patient care. Using the mobile package enables me to accomplish tasks more quickly. Using the mobile package in my job increases my productivity. Using the mobile system makes it easier to do my job.

12345 12345 12345 12345

11 Ease of use Learning to operate the mobile package is easy for me. I find the mobile package easy to use. It is easy for me to become a skilful user of the mobile package. I find the mobile package inflexible to interact with.

12345 12345 12345 12345

12. Compatibility Using the mobile package fits into my work style. I think that using the mobile package fits well with my life style. Using the mobile package helps me in my working routines.

12345 12345 12345

13. System current usage 13.1 Usage frequency: On the average, I use the mobile package: 1. I don’t use it at all. 2. About once a month. 3. About once a week. 4. Several times a week. 5. About once a day. 6. Several times a day.

13.2 Usage Volume: Please specify (estimate) how many hours each week you normally spend using the mobile package? 1.